Zepplin: truncated output with self defined sql interpreter

I use Zeppelin 0.7.2 and created for testing purposes an own "interpreter" for directly querying a sql database using jdbc. If I use the sql query select col from table in zeppelin, check the scatter plot view or download the data as csv file from the inbuilt view I get only about 7.800 rows. The interpreter is configured with common.max_count=1000000000 and zeppelin.interpreter.output.limit=102400

If I use the same query with a sql client I get about 22.000 row.

So I wonder why there is a difference.

Solution

I solved it by modify the config file : conf/zeppelin-site.xml

raise the size of :

zeppelin.interpreter.output.limit

zeppelin.websocket.max.text.message.size